floodway provided therein unless the work has been recommended by the Chief of Engineers and authorized by the Secretary of War.

(b) Providing that the penalties and procedure applicable to violations of the laws for the protection and preservation of the navigable waters of the United States, enacted in Sections 12 and 17 of the River and Harbor Act of March 3, 1899, shall apply to violations of the above provision of law.

(c) Providing that existing laws relating to the acquisition of lands, easements on rights of way needed for a work of river and harbor improvement shall be applicable to the acquisition of lands, easements or rights of way for flood control works.

(d) Amending Sections 3 and 4 of the Act of June 28, 1879, constituting the Mississippi River Commission; to provide that it shall be the duty of said Commission to advise on all questions relating to the improvement of navigation on the Missisippi River and the prevention of destructive floods which may be referred to the Commission by the President of the Commission or higher authority, and to provide that the President of the Mississippi River Commission shall have the qualifications now prescribed by law for the Assistant Chief of Engineers, and shall receive the rank, pay and allowances of a Brigadier General of Engineers while actually assigned to such duty by competent orders.

N 1879 the Government of the United States recognized that the control of the periodic floods of the Mississippi River was a national problem. To meet this problem Congress created the Mississippi River Commission, consisting of seven members, three of whom are selected from the Corps of Engineers, United States Army, one from the Coast and Geodetic Survey, and three from civil life. This Commission operates under the direction of the Secretary of War and the Chief of Engineers. Its integrity and devotion to duty have never been questioned.

This Commission after a careful study developed a plan for controlling the floods by means of a system of levees and, with the approval of Congress, work thereon started in 1882. Up to June 30, 1926, there had been expended in the furtherance of this plan a total of $174,469,931.06 on the Mississippi River, $81,686,285.56, including $16,689,146.66 contributed by local interests, being for flood control.

In his report for the year ending June 30, 1926, the Chief of Engineers, United States Army (page 1793), said:

Effect of Improvement.-It may be stated in a general way, the improvement is providing a safe and adequate channel for navigation and is now in condition to prevent the destructive effect of floods.

Then came the flood of 1927, the greatest physical disaster in American history. Of the 30,000 square miles. threatened with inundation, the levees successfully protected 12,000 square miles having a larger percentage of

cultivated land than the 18,000 square miles flooded. Nevertheless the resulting loss of property was great and fell upon a people already heavily debtburdened for the construction of costly flood defense works. Loss of human life was providentially small, but the suffering, misery, and despair was such as is only known by a people who have to flee their homes before the advancing hosts of an invading army. In this case the invading army was water instead of armed men. The element of personal conflict was absent but in its stead was an impersonal relentless element that covered all things, penetrated everywhere, and when it withdrew left behind only death, filth, ruin and decay.

These facts are recited, not with the thought of charging delinquency to the Commission. Omnipotence cannot be expected of any human agency. Rather are they recalled as proof that the problem is one of controlling great natural forces and that no system of protection that man erects can have factors of safety so great that assistance from other sources should be ignored. Doubtless the Commission itself would be the first to subscribe to such a conclusion. Probably there is nowhere in the country today any considerable sentiment for "levees only," but even the writer of this article acknowledges a conviction favoring "levees first."

In the past there has probably been a tendency to overstate the case both for and against forests as a factor in flood control. Let it, therefore, be understood at the outset that the writer does not advocate forests as a substi

tute for levees. Levees are necessary now, always have been necessary, and certainly will be necessary for generations to come. They should be constructed high enough and strong enough to control the greatest flood within the range of reasonable probability. This country is rich enough and resourceful enough to construct a system of levees of that kind, and the writer would not advocate diverting from such a project a single dollar necessary to its successful completion.

The problem of providing immediate protection to the region threatened by flood waters can only be solved by the construction engineer whose field of direct responsibility is the protection of a limited area of about 30,000 square miles, formerly used by the Mississippi River periodically as part of a great flood channel, but now taken over by man for agricultural and other purposes. At the same time there is a great problem of land use which also has a direct bearing on the regimen of the Mississippi. It involves the entire area of about 1,225,000 square miles which contributes its surplus waters to that stream. The use which is made of any part of that vast area directly affects runoff and runoff makes floods.

One of the most common forms of use to which land is put is the growing of forests. Generally speaking, the lands so used are those having the most direct bearing on floods. They are lands too rough and steep for cultivation and the high mountain lands which receive a heavier precipitation than the plains lands. Forests of swamp or bottom type and on level sandy soils which absorb the heaviest rainfall without washing are exceptions to the general rule.

Originally about 60 per cent of the Mississippi basin was treeless and 40 per cent forest. In De Soto's time the river was subject to floods. The

Missouri brought down its great loads of silt from the "Badlands" and the Arkansas brought its burden of soil from "The Breaks" at the edge of the "Staked Plains," and each spring the flood waters deposited a layer of silt over the 30,000 square miles which we now wish to permanently reclaim.

Since De Soto's day the proportion of forest land in the valley has been cut to 20 per cent, or about 250,000 square miles. It is idle to expect that the proper use of this relatively small proportion of the entire watershed could prevent floods. Upon the other hand it is equally absurd to say that the use of so great an area has no bearing on resulting floods and runoff. Avoiding extremes in either case, one may expect provisions for the efficient protection and management of the forests now existing on the watershed and the replanting of lands which are now washing badly and should never have been cleared, will be included in future plans for Mississippi flood control, and that such use of land will add materially to the security and permanent usefulness of the supplementary levees and other necessary engineering structures.

For a clear understanding of the influence of forests on streamflow regulation one must take into consideration certain laws of physics governing the power of running water. First, the erosive or abrading power of a stream varies as the square of its velocity. For example, if the velocity of the stream is increased ten times its erosive power is increased one hundred times. Second, the transporting power of water varies as the sixth power of its velocity. For example, if the velocity of a stream is increased ten times, its transporting power is increased one million times. A current two miles an hour will move along its channel fragments of stone the size of a hen's egg, weighing about three ounces, while a

torrent of twenty miles an hour will carry boulders weighing one hundred

tons.

It is a self-evident fact that water falling on steep lands tends to rapid runoff. The laws just cited show that erosion is tremendously increased by rapidity of runoff. It follows that for any given type of soil the steepest lands are most subject to erosion. Therefore it is upon steep rough lands that forests as an erosion-preventive factor are most important.

Coincident with retarding erosion and runoff, forests also retain a considerable volume of moisture and increase the amount taken up by the soil. Not only is the earth protected from beating rainfall but the precipitation is broken up by the leaf canopy, and must either settle in the form of mist or reassemble on twigs and branches and slowly trickle to the ground. Also, if the forest has been protected from fire and excessive tramping of stock, the trickling moisture encounters a layer of humus composed of dead leaves and other vegetation which readily absorbs water amounting to several hundred times its weight.

A layer of humus four inches deep will easily retain and transmit to the soil one inch of rainfall. This statement is conservative. The immediate water storage capacity of such a forest may therefore be calculated as equivalent to 3630 cubic feet. The effect of increasing water storage on a single acre would hardly be noticeable on the smallest stream. The effect of even a million acres would scarcely be noticed in the volume of a great Mississippi flood. However, it is just as true of the Mississippi flood as it is of the smallest rivulet that its burden of liquid and solids is made up of little drops of water and little grains of land. The total area of forest land and lands once forested but now unused and

bare of forest growth in the Mississippi drainage basin is 250,000 square miles. About 200,000 square miles of such forests and forest land is far from doing what it might to hold back flood waters and prevent erosion. Some lands are in fairly good shape; some very bad shape. To rate the present water holding capacity of any but the best of these lands at 50 per cent that of a well stocked, well protected forest is generous indeed. If we reforest the bare unused forest lands and protect the remaining forests, both old and young, from fires, and prevent overgrazing, this should on the average increase the layer of humus on 200,000 square miles by two inches. Such an increase means 1815 cubic feet of water storage by each acre of forest land so improved. Applied to 200,000 square miles such improved forest conditions would give an increased water storage capacity totalling 232,320,000,000 cubic feet, or the equivalent of the entire reported average flow of the Mississippi River at Quincy, Illinois, for a period of nearly thirty-seven days.

For the purpose of assisting in the regulation of streamflow on the Mississippi, dams have been constructed at the outlet of Lake Winnebigoshish and five other lakes at the headwaters of that river, thereby creating what is said to be the greatest water impounding system in the world. These six great storage reservoirs have a total impounding capacity of somewhat over ninety-five billion cubic feet. The forests of the valley can be made to store and regulate the flow of about two and a half times the total capacity of these great reservoirs over and above the present water storage service, while at the same time increasing their value and effectiveness for wood production, outdoor recreation, and other purposes. This will not prevent floods but will reduce them.

However, water storage is not the greatest service performed by forests in the direction of flood control. More important than water storage is their influence in preventing soil erosion, which after all is the problem most vital to the Mississippi Valley. Floods in the lower valley threaten periodically a region embracing a total of 30,000 square miles, but soil erosion is nibbling constantly at the fertility of the 1,225,000 square miles of watershed contributing to flood runoff.

The whole world knows that the Mississippi flood of 1927 temporarily inundated a total of about 18,000 square miles, much of it cultivated. But few know or care that because of soil erosion 1842 square miles of plowed land in four small tributary drainage units just east of the bottom lands in Mississippi was abandoned between 1920 and 1925. Combined, the four units had a gross area of only 13,336 square miles. The abandoned plowed land of the five year period was therefore over 13 per cent of the gross area. In one unit 50 per cent of the plowed land was abandoned within the two dates.

These cases are exterme examples of what is taking place in many parts of the Mississippi watershed. In the aggregate it is reasonably certain that a greater acreage of farm land has had its value for cultivation permanently destroyed by soil erosion during the past five years by floods on the Mississippi watershed than the gross area endangered by inundation from Cairo to the Gulf.

Undoubtedly more poverty and suffering have resulted from soil erosion than from floods, but it creates no thrilling spectacle, gets no newspaper headlines and no Red Cross relief. And yet the family that is driven. from its home by gully washing is

a flood sufferer whose misery is no less acute because comparatively unknown.

Land with erodible soils having a slope so slight as 334 per cent will under the ordinary rainfall conditions of the great agricultural sections of the Mississippi Valley lose on the average, either as bare or cultivated land, twentythree tons of soil per acre annually. The cultivation of lands with a much greater gradient, even up to two or three times as great, is not uncommon, with erosion increasing at a greatly accelerated rate.

Silt in a stream not only increases its volume but also multiplies its power to destroy. It gives it an abrasive capacity not possessed by a clear stream. Forests are the best means of checking erosion. Streams originating in timber-covered hills are clear and pure.

Forests cannot remove from the Mississippi all its great burden of silt, amounting annually to somewhere between half a billion and a billion tons. They can, however, if maintained on steep lands largely reduce this silt burden as well as assisting in regulating the streamflow. Forests cannot prevent floods on the Mississippi, but can be made to materially increase the safety factor of levees and other protection works. Reforestation would also extend the periods of usefulness of such works. It is a self-evident fact that a balance must eventually be established and maintained between the amount of silt annually received by the river between the levees below Cairo and the amount discharged into the Gulf of Mexico. If the amount received greatly exceeds the amount discharged into the Gulf of Mexico, the day of flood destruction must come sooner or later. It is cheaper and better to keep rich soil on hillside slopes by growing forests than to per